RE: [Xen-devel] [PATCH][XEND]Fix memory allocation for VTi domain with new Qemu on xen-unstagle.hg

["Zhang, Xiantao" <xiantao.zhang@xxxxxxxxx>]

Hi Keir,
        Thank you for your comments. This should be a temporary solution for 
IA64 side. Currently, VTi assumes domain's total pages == max_pages at creation 
time. In addition, VTi guest doesn't support maxmem option in configuration 
file. Current code will assume domain's maxmem == domain's memory which doesn't 
include guest FW and xenstore and IO share page when no option for maxmem in 
configuration file. And XC will call domain_memory_increase_reservation to 
reserve memory for domain with a large memory value which includes domain's 
memory, guest FW size  and IO share page, so hypervisor will complain no more 
memory can be assigned to this domain. Once balloon driver is ready @IA64 side, 
we will handle it like IA32. Can it be a workaround now? 
Thanks & Best Regards
-Xiantao

OTC,Intel Corporation

> -----Original Message-----
> From: Keir Fraser [mailto:Keir.Fraser@xxxxxxxxxxxx]
> Sent: 2006年7月25日 18:23
> To: Zhang, Xiantao
> Cc: xen-devel@xxxxxxxxxxxxxxxxxxx; xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> Subject: Re: [Xen-devel] [PATCH][XEND]Fix memory allocation for VTi domain
> with new Qemu on xen-unstagle.hg
> 
> 
> On 25 Jul 2006, at 02:59, Zhang, Xiantao wrote:
> 
> > Due to IA64 balloon driver not ready and it depends on max memory value
> > to allocate its memory. So this fix is necessary now.
> > Thanks & Best Regards
> > -Xiantao
> 
> Can you explain how introduction of new qemu code has made this change
> necessary? I'm surprised it would be affected by maxmem setting,
> particularly in a different way to the older qemu code we were using
> previously.
> 
>   -- Keir

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel